Package de.jungblut.classification.knn

Class AbstractKNearestNeighbours

java.lang.Object

de.jungblut.classification.AbstractPredictor

de.jungblut.classification.AbstractClassifier

de.jungblut.classification.knn.AbstractKNearestNeighbours

All Implemented Interfaces:

Classifier, Predictor

Direct Known Subclasses:

KNearestNeighbours, SparseKNearestNeighbours

public abstract class AbstractKNearestNeighbours
extends AbstractClassifier

K nearest neighbour classification algorithm that is seeded with a "database" of known examples and predicts based on the k-nearest neighbours majority vote for a class.

Field Summary

Fields

Modifier and Type	Field	Description
protected int	k
protected int	numOutcomes

Constructor Summary

Constructors

Constructor	Description
AbstractKNearestNeighbours(int numOutcomes, int k)	Constructs a new knn classifier.

Method Summary

Modifier and Type	Method	Description
protected abstract java.util.List<de.jungblut.jrpt.VectorDistanceTuple<de.jungblut.math.DoubleVector>>	getNearestNeighbours(de.jungblut.math.DoubleVector feature, int k)	Find the k nearest neighbours for the given feature.
de.jungblut.math.DoubleVector	predict(de.jungblut.math.DoubleVector features)	Classifies the given features.
de.jungblut.math.DoubleVector	predictProbability(de.jungblut.math.DoubleVector features)	Classifies the given features.

Methods inherited from class de.jungblut.classification.AbstractClassifier

train, train

Methods inherited from class de.jungblut.classification.AbstractPredictor

extractPredictedClass, extractPredictedClass, predictedClass, predictedClass

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface de.jungblut.classification.Predictor

asClassifier, extractPredictedClass, extractPredictedClass, predictedClass, predictedClass

Field Detail

numOutcomes

protected final int numOutcomes

k

protected final int k

Constructor Detail

AbstractKNearestNeighbours

public AbstractKNearestNeighbours(int numOutcomes,
                                  int k)

Constructs a new knn classifier.

Parameters:

numOutcomes - the number of different outcomes that can be predicted.

k - the number of neighbours to analyse to get a prediction (it does so by majority voting).

Method Detail

predict

public de.jungblut.math.DoubleVector predict(de.jungblut.math.DoubleVector features)

Description copied from interface: Predictor

Classifies the given features.

Returns:

If the number of outcomes is 2 (binary prediction) the returned vector contains the class id (0 or 1) at the first index. If not, a histogram of the classes that were predicted.

predictProbability

public de.jungblut.math.DoubleVector predictProbability(de.jungblut.math.DoubleVector features)

Description copied from interface: Predictor

Classifies the given features.

Specified by:

predictProbability in interface Predictor

Overrides:

predictProbability in class AbstractPredictor

Returns:

a vector that returns the probability of all outcomes. The output vector should sum to one.

getNearestNeighbours

protected abstract java.util.List<de.jungblut.jrpt.VectorDistanceTuple<de.jungblut.math.DoubleVector>> getNearestNeighbours(de.jungblut.math.DoubleVector feature,
                                                                                                                            int k)

Find the k nearest neighbours for the given feature.

Parameters:

feature - the feature to find neighbours for.

k - the number of neighbours to find.

Returns:

a list of VectorDistanceTuple's that contain the outcome of the retrieved vectors.